On Thu, 21 Dec 2023 15:21:08 GMT, Scott Gibbons <sgibb...@openjdk.org> wrote:
>> Re-write the IndexOf code without the use of the pcmpestri instruction, only
>> using AVX2 instructions. This change accelerates String.IndexOf on average
>> 1.3x for AVX2. The benchmark numbers:
>>
>>
>> Benchmark Score
>> Latest
>> StringIndexOf.advancedWithMediumSub 343.573 317.934
>> 0.925375393x
>> StringIndexOf.advancedWithShortSub1 1039.081 1053.96
>> 1.014319384x
>> StringIndexOf.advancedWithShortSub2 55.828 110.541
>> 1.980027943x
>> StringIndexOf.constantPattern 9.361 11.906
>> 1.271872663x
>> StringIndexOf.searchCharLongSuccess 4.216 4.218
>> 1.000474383x
>> StringIndexOf.searchCharMediumSuccess 3.133 3.216
>> 1.02649218x
>> StringIndexOf.searchCharShortSuccess 3.76 3.761
>> 1.000265957x
>> StringIndexOf.success 9.186
>> 9.713 1.057369911x
>> StringIndexOf.successBig 14.341 46.343
>> 3.231504079x
>> StringIndexOfChar.latin1_AVX2_String 6220.918 12154.52
>> 1.953814533x
>> StringIndexOfChar.latin1_AVX2_char 5503.556 5540.044
>> 1.006629895x
>> StringIndexOfChar.latin1_SSE4_String 6978.854 6818.689
>> 0.977049957x
>> StringIndexOfChar.latin1_SSE4_char 5657.499 5474.624
>> 0.967675646x
>> StringIndexOfChar.latin1_Short_String 7132.541
>> 6863.359 0.962260014x
>> StringIndexOfChar.latin1_Short_char 16013.389 16162.437
>> 1.009307711x
>> StringIndexOfChar.latin1_mixed_String 7386.123 14771.622
>> 1.999915517x
>> StringIndexOfChar.latin1_mixed_char 9901.671 9782.245
>> 0.987938803
>
> Scott Gibbons has updated the pull request incrementally with one additional
> commit since the last revision:
>
> Addressing review comments.
src/hotspot/share/opto/library_call.cpp line 1202:
> 1200:
> 1201: Node* result = nullptr;
> 1202: bool do_intrinsic =
Name change suggestion: do_intrinsic -> call_opt_stub
src/hotspot/share/opto/library_call.cpp line 1224:
> 1222: result = _gvn.transform(new ProjNode(call, TypeFunc::Parms));
> 1223: } else {
> 1224: result = make_indexOf_node(src_start, src_count, tgt_start,
> tgt_count,
Existing routines emits IR to handle following special cases
1) tgt_cnt > src_cnt return -1
2) tgt_cnt == 0 return 0.
Should we not be preserving those check before calling stub ?
src/hotspot/share/opto/library_call.cpp line 1273:
> 1271: Node* result = nullptr;
> 1272:
> 1273: if ((StubRoutines::string_indexof() != nullptr) && (ae ==
> StrIntrinsicNode::LL)) {
Why are we not calling stub for StrIntrinsicNode::UU
-------------
PR Review Comment: https://git.openjdk.org/jdk/pull/16753#discussion_r1444390460
PR Review Comment: https://git.openjdk.org/jdk/pull/16753#discussion_r1444406814
PR Review Comment: https://git.openjdk.org/jdk/pull/16753#discussion_r1444420392
